canonical

A canonical is a <link rel="canonical" href="..."> element in a page's head, or an equivalent HTTP header, naming the single URL that should represent that content. It solves a common problem: the same page is reachable with and without a trailing slash, with tracking parameters, through http and https, or with www and without, and crawlers would otherwise treat each as a separate, duplicated page.

For search, correct canonicals consolidate signals onto one URL and prevent duplicate-content dilution. For AI retrieval they do a related job: they help an index settle on one address for your service page so citations point to the clean URL rather than a parameterised variant, and so a model does not see two near-identical pages and treat them as conflicting sources.

Common faults, all of which Seoptist's technical audit checks for, include:

• Missing canonicals on key pages.
• Canonicals pointing to a redirecting, 404 or noindex URL.
• Every page canonicalising to the homepage.
• Canonicals that disagree with the sitemap or with hreflang annotations.
• Relative URLs where absolute ones are expected.

Each fault appears in the checklist with the affected URLs and the exact tag to use.